Hello Everyone! We apologize for moving this again but we have found a location that will host us. We look forward to seeing you there!
The promise of big data is better predictions. There is no best model that works for all of your data. Model predictive performance is domain specific. What works in one data domain has sometimes very little consequence in another one. Data science needs to get closer to the business and unlock value.
Ensembles are here to stay! Users want a buffet of algorithms that try to "lock-pick" the data for it's secrets. Time is eventually the key limiter. Data science efforts have to make best out of the budget for experimentation and use some kind of co-evolutionary technique that picks the "Champion" model of models for your data. Robust automation and fast analytics can speedup large parts of data smithy. In this talk we discuss ensemble techniques of boosting & trees that when applied on use cases lead to a substantial better predictions. H2O Open Source Machine Learning Platform will be used as a demo bed for GBM and RF.